The long-term goal is to understand how innate predispositions and experience interact in the acquisition of a complex, learned behavior: bird song. Bird song is the only known non-human model for complex vocal learning, which is thought to involve both innate predispositions and experience in humans as well. Specific aims of the proposed research are to investigate whether innately recognized acoustic cues guide vocal learning, whether different cues act at different stages of development, and how song recognition changes over development and with experience, on both behavioral and neural levels. The white-crowned sparrow, Zonotrichia leucophrys nuttalli, will be used as the subject in three experiments: (1) birds will be tutored with an array of modified songs designed to identify which acoustic properties are preferentially learned, (2) vocal recognition tests will be done on birds preceding and following tutoring, to determine how tutor exposure affects song recognition, and (3) neural responses to playback of various song stimuli will be measured in the brain nucleus HVC after birds have learned abnormal tutor songs, to investigate the degree to which the neural representation of song is modified by experience.